DAAO venues and organisations¶

This is an exploratory data analysis of collected data from DAAO. We focus on venues and organisations adopting natural language processing to reveal patterns in the data.

The visualisations consist of…

word clouds
dendrograms
time series and temporal bar charts

import pandas as pd
import numpy as np
import json
from collections import Counter
import plotly.express as px
import plotly.graph_objects as go

# dfs = pd.read_csv('data/acde_event_202304010318.csv', low_memory=False)

# dfs_expanded = []

# # get start and end years
# for idx,row in dfs[dfs.data_source == 'DAAO'].iterrows():
#     try: 
#         this_locations = pd.json_normalize(eval(row['locations'].replace("ObjectId('","'").replace("')","'")))

#         for idx2, row2 in this_locations.iterrows():
#             try: start_yr = row2['coverage_range.date_range.date_start.year']
#             except: start_yr = None

#             try: end_yr = row2['coverage_range.date_range.date_end.year']
#             except: end_yr = None

#             try: place_address = row2['coverage_range.place.ori_address']
#             except: place_address = None

#             try: 
#                 latitude = row2['coverage_range.place.geo_coord.latitude']
#                 longitude = row2['coverage_range.place.geo_coord.longitude']
#             except:
#                 latitude = None; longitude = None

#             row['start_year'] = int(start_yr); row['end_year'] = int(end_yr)
#             row['latitude'] = latitude; row['longitude'] = longitude
#             row['place_address'] = place_address
#             dfs_expanded.append(row)

#     except:
#         start_yr = None; end_yr = None; latitude = None; longitude = None; place_address = None
#         dfs_expanded.append(row)

# pd.DataFrame(dfs_expanded).to_csv('data/acde_event_expanded.csv', index=False)

# read in expanded data
dfs_expanded = pd.read_csv('data/acde_event_expanded.csv', low_memory=False)

locations_cond = dfs_expanded.latitude.notnull()
year_cond = dfs_expanded.start_year.notnull()
desc_cond = dfs_expanded.description.notnull()
period_cond = (dfs_expanded.start_year >= 1900) & (dfs_expanded.start_year <= 2020)

print('In DAAO,') 
print(f'- there are {dfs_expanded.shape[0]} events.')
print(f'- there are {dfs_expanded[year_cond].shape[0]} events with dates.')
print(f'- there are {dfs_expanded[locations_cond & year_cond].shape[0]} events with geocodes and dates.')
print(f'- there are {dfs_expanded[locations_cond & year_cond & desc_cond].shape[0]} events with geocodes, dates and biographical information.')
print(f'- there are {dfs_expanded[locations_cond & year_cond & desc_cond & period_cond].shape[0]} events with geocodes, dates and biographical information after filtering for events between 1900 and 2020.')

In DAAO,
- there are 22524 events.
- there are 21242 events with dates.
- there are 17764 events with geocodes and dates.
- there are 7751 events with geocodes, dates and biographical information.
- there are 7581 events with geocodes, dates and biographical information after filtering for events between 1900 and 2020.

dfs_rich = dfs_expanded[locations_cond & year_cond & desc_cond & period_cond].copy()

type_count = dfs_rich['types'].value_counts().reset_index().rename(columns={'index':'type','types':'count'})
type_count['prop'] = round((type_count['count']/dfs_rich.shape[0])*100,2)
exh_prop = type_count['type'].str.contains('exhibition')
other_prop = type_count['type'].str.contains('other event')
fest_prop = type_count['type'].str.contains('festival')
perf_prop = type_count['type'].str.contains('performance')

# # create a decade column
dfs_rich['decade_start'] = [str(int(x))[:3]+'0' for x in dfs_rich['start_year']]
dfs_rich['decade_start'] = dfs_rich['decade_start'].astype(int)

print('\nWe use the 7,581 events for the remainder of the analysis. In this subset,') 
print(f'- there are {type_count[exh_prop]["count"].sum()} exhibitions ({round(type_count[exh_prop]["prop"].sum(),2)}%).')
print(f'- there are {type_count[other_prop]["count"].sum()} other-events ({round(type_count[other_prop]["prop"].sum(),2)}%).')
print(f'- there are {type_count[fest_prop]["count"].sum()} festivals ({round(type_count[fest_prop]["prop"].sum(),2)}%).')
print(f'- there are {type_count[perf_prop]["count"].sum()} performance-event ({round(type_count[perf_prop]["prop"].sum(),2)}%).')
print(f'- there are {dfs_rich["types"].isnull().sum()} events with missing type data ({round((dfs_rich["types"].isnull().sum()/dfs_rich.shape[0])*100,2)}%).')

print('\nIn terms of time,')
for t in range(1900,2021,10):
    if t != 2020: print(f'- there are {dfs_rich[dfs_rich.decade_start == t].shape[0]} events between {t}-{t+10} ({round((dfs_rich[dfs_rich.decade_start == t].shape[0]/dfs_rich.shape[0])*100,2)}%).')
    else: print(f'- there are {dfs_rich[dfs_rich.start_year == t].shape[0]} events in {t} ({round((dfs_rich[dfs_rich.start_year == t].shape[0]/dfs_rich.shape[0])*100,2)}%).')

We use the 7,581 events for the remainder of the analysis. In this subset,
- there are 6940 exhibitions (91.55%).
- there are 86 other-events (1.13%).
- there are 32 festivals (0.42%).
- there are 21 performance-event (0.28%).
- there are 508 events with missing type data (6.7%).

In terms of time,
- there are 33 events between 1900-1910 (0.44%).
- there are 31 events between 1910-1920 (0.41%).
- there are 39 events between 1920-1930 (0.51%).
- there are 40 events between 1930-1940 (0.53%).
- there are 41 events between 1940-1950 (0.54%).
- there are 90 events between 1950-1960 (1.19%).
- there are 159 events between 1960-1970 (2.1%).
- there are 651 events between 1970-1980 (8.59%).
- there are 1413 events between 1980-1990 (18.64%).
- there are 2272 events between 1990-2000 (29.97%).
- there are 2283 events between 2000-2010 (30.11%).
- there are 526 events between 2010-2020 (6.94%).
- there are 3 events in 2020 (0.04%).

Most frequent geocodes¶

Before jumping into the visuals, we inspect the nuances of the geocodes attached to exhibtion data. We find that the geocodes are not always accurate, and that there are many distinct venues representing the same geocode. Further pre-processing will need to be conducted to ensure downstream trends can be accurately identified.

Below is a list of the top 20 geocodes, and the number of events they represent. Through further inspection we can see which geocodes accurate represent the place names.

most_freq_geocodes = dfs_rich[['latitude','longitude']].value_counts().reset_index().rename(columns={0:'Frequency'})
most_freq_geocodes = most_freq_geocodes.head(20)

# get the most frequent place_address for each geocode
for idx,row in most_freq_geocodes.iterrows():
    place_address = dfs_rich[(dfs_rich.latitude == row['latitude']) & (dfs_rich.longitude == row['longitude'])]['place_address'].value_counts().index[0]

    try: place_address2 = dfs_rich[(dfs_rich.latitude == row['latitude']) & (dfs_rich.longitude == row['longitude'])]['place_address'].value_counts().index[1]
    except: place_address2 = None
    
    try: place_address3 = dfs_rich[(dfs_rich.latitude == row['latitude']) & (dfs_rich.longitude == row['longitude'])]['place_address'].value_counts().index[2]
    except: place_address3 = None

    try: place_address4 = dfs_rich[(dfs_rich.latitude == row['latitude']) & (dfs_rich.longitude == row['longitude'])]['place_address'].value_counts().index[3]
    except: place_address4 = None

    try: place_address5 = dfs_rich[(dfs_rich.latitude == row['latitude']) & (dfs_rich.longitude == row['longitude'])]['place_address'].value_counts().index[4]
    except: place_address5 = None
    
    most_freq_geocodes.loc[idx,'most_frequent_address'] = place_address
    most_freq_geocodes.loc[idx,'2ndmost_frequent_address'] = place_address2
    most_freq_geocodes.loc[idx,'3rdmost_frequent_address'] = place_address3
    most_freq_geocodes.loc[idx,'4thmost_frequent_address'] = place_address4
    most_freq_geocodes.loc[idx,'5thmost_frequent_address'] = place_address5

most_freq_geocodes

	latitude	longitude	Frequency	most_frequent_address	2ndmost_frequent_address	3rdmost_frequent_address	4thmost_frequent_address	5thmost_frequent_address
0	-34.928621	138.599959	936	Experimental Art Foundation, Adelaide, SA	Contemporary Art Centre of South Australia, Ad...	Adelaide Central Gallery, Adelaide, SA	Greenaway Art Gallery, Adelaide, SA	Flinders University Art Museum, Adelaide, SA
1	-33.868901	151.207091	898	Artspace, Sydney, NSW	Sydney, NSW	Sherman Galleries, Sydney, NSW	Australian Centre for Photography, Sydney, NSW	Ivan Dougherty Gallery, Sydney, NSW
2	-37.813187	144.962980	502	Monash University Museum of Art, Melbourne, VIC	Australian Centre for Contemporary Art, Melbou...	RMIT Gallery, Melbourne, VIC	Karen Woodbury Gallery, Melbourne, Vic.	Melbourne, Vic.
3	-27.470933	153.023502	494	Institute of Modern Art, Brisbane, QLD	Brisbane City Art Gallery, Brisbane, Qld	Brisbane, Qld	University of Queensland Art Museum, Brisbane,...	Brisbane, QLD
4	-33.868627	151.217062	321	Art Gallery of New South Wales, Sydney, NSW	Art Gallery of NSW, Sydney, NSW	Art Gallery Of New South Wales, Sydney, NSW	[National] Art Gallery of New South Wales, Syd...	Art Gallery of New South Wales, Sydney, 14 Oct...
5	-34.416667	150.883333	192	Wollongong City Gallery, Wollongong, NSW	Wollongong City Art Gallery, Wollongong, NSW	Project Contemporary Artspace, Wollongong, NSW	Wollongong City Gallery, NSW	Project Centre for Contemporary Art, Wollongon...
6	-12.437368	130.833821	152	Museums and Art Galleries of the Northern Terr...	Museum and Art Gallery of the Northern Territo...	Museum & Art Gallery of the Northern Territory...	Museum and Art Gallery of the Northern Territo...	Museums and Art Galleries of the Northern Terr...
7	-37.822536	144.969113	134	National Gallery of Victoria, Melbourne, VIC	National Gallery of Victoria, Melbourne, Vic.	Ian Potter Centre, National Gallery of Victori...	Access Gallery, National Gallery of Victoria, ...	National GAllery of Victoria, Melbourne, Vic.
8	-27.472096	153.018173	133	Queensland Art Gallery, Brisbane, QLD	Queensland Art Gallery, Brisbane, Qld	Queensland Art Gallery, Brisbane, Qld.	Queensland Art Gallery, Brisbane. QLD	Queensland Art Gallery, Brsbane, QLD
9	-32.929205	151.772973	126	Newcastle Region Art Gallery, Newcastle, NSW	Newcastle City Art Gallery, Newcastle, New Sou...	Newcastle Regional Art Gallery, Newcastle, NSW	Newcastle Regional Art Gallery, NSW	None
10	-31.952222	115.858889	117	John Curtin Gallery, Curtin University, Perth, WA	Perth, WA	Perth Institute of Contemporary Art, Perth, WA	Greenhill Galleries, Perth, WA	Katanning Arts Centre, Katanning & Western Aus...
11	-33.281886	149.102659	109	Orange Regional Gallery, Orange, NSW	Orange Regional Art Gallery, Orange, NSW	Orange Regional Gallery, NSW	Orange Regional Gallery	None
12	-38.385530	142.482454	108	Warrnambool Art Gallery, Warrnambool, VIC	Warrnambool Art Gallery, VIC	None	None	None
13	-31.950414	115.860604	106	Art Gallery of Western Australia, Perth, WA	Art Gallery of Western Australia, Perth, WA (a...	None	None	None
14	-33.872355	151.224854	103	8 Llankelly Place, Kings Cross, Sydney, 2011	None	None	None	None
15	-41.437456	147.134155	95	Queen Victoria Museum and Art Gallery, Launces...	Queen Victoria Museum and Art Gallery, Launces...	Queen Victoria Museum and Art Gallery, Launces...	None	None
16	-33.884611	151.226586	94	Ivan Dougherty Gallery, COFA, UNSW, Paddington...	Roslyn Oxley9 Gallery, Paddington, Sydney, NSW	Mary Place Gallery, Paddington, NSW	Kaliman Gallery, Paddington, Sydney, NSW	Global Gallery, Paddington, NSW
17	-37.562107	143.856149	92	Ballarat Fine Art Gallery, Ballarat, VIC	Ballarat, VIC	Ballarat Fine Art Gallery, Ballarat, Vic.	None	None
18	-33.888493	151.187100	86	Tin Sheds Gallery, University of Sydney, Sydne...	Sydney College of the Arts, University of Sydn...	Tin Sheds, University of Sydney, New South Wales	Tinsheds Gallery, University of Sydney, Sydney...	SCA Galleries, Sydney College of the Arts, Uni...
19	-37.758900	145.084301	85	Heide Museum of Modern Art, Melbourne, VIC	Heide Museum of Modern Art, Bulleen, Vic.	Heide Museum of Modern Art, VIC	Heide Museum of Modern Art, Bulleen, VIC	None

Top 10 geocoded places with most events¶

After omitting geocodes with random or erroneous place names, we can see that the “true” top 10 geocoded places with the most events. As expected these are all art galleries - all with over 100 events.

Venue	City	Frequency
Art Gallery of New South Wales	Sydney, NSW	321
Wollongong City Gallery	Wollongong, NSW	192
Museums and Art Galleries of the Northern Territory	Darwin, NT	152
National Gallery of Victoria	Melbourne, VIC	134
Queensland Art Gallery	Brisbane, QLD	133
Newcastle Region Art Gallery	Newcastle, NSW	126
Orange Regional Gallery	Orange, NSW	109
Warrnambool Art Gallery	Warrnambool, VIC	108
Art Gallery of Western Australia	Perth, WA	106
8 Llankelly Place, Kings Cross	Sydney, NSW	103

Most frequent terms used for place names¶

From here, we explore the most frequent terms used for place names. We find that the most frequent terms are “Gallery”, “Art”, “Centre”, “Museum”, and “University”.

# find the most frequent words across the strings for place names
from collections import Counter
import matplotlib.pyplot as plt

import nltk
# nltk.download('punkt')

from nltk.tokenize import word_tokenize

# create a list of all the place names
all_place_names = dfs_rich['place_address'].astype(str).tolist()

# create a list of all the words in the place names
all_words = []
for place_name in all_place_names:
    all_words.extend(word_tokenize(place_name))

# find top 100 most frequent words
most_freq_words = Counter(all_words).most_common(89)

# remove a list of all words that are not relevant
words_to_remove = [',','NSW','Sydney','Melbourne','Adelaide','South','SA','Brisbane','VIC',
                   '.', 'New', 'Australia', 'QLD', 'Vic', 'Wales', 'WA', 'Canberra', 'and',
                   'Perth', 'ACT', 'of', 'Qld', 'Victoria','Wollongong','TAS','Queensland','Newcastle',
                   'Street','Hobart','the','The','Launceston','Orange','UK','NT','London','USA',
                   'Paddington','Darwin','for','Western','Warrnambool','Ballarat','Northern','Territory',
                   'England','Watters','Macquarie','Artspace','St',"'s",'&','Potter','Kings','Ian','Cross',
                   '8','Llankelly','2011','Fremantle','Queen','Ivan','Dougherty','Tasmania','Central',
                   'Curtin','France','Tin','Sheds','York','Monash','Paris','Heide','Tasmanian']

# remove the words from the list of most frequent words
most_freq_words = [word for word in most_freq_words if word[0] not in words_to_remove]

most_freq_words_dict = dict(most_freq_words)

# add value of two keys
most_freq_words_dict['Gallery'] = most_freq_words_dict['Gallery'] + most_freq_words_dict['Galleries']
most_freq_words_dict['Museum'] = most_freq_words_dict['Museum'] + most_freq_words_dict['Museums']

# remove key 'Gallery'
most_freq_words_dict.pop('Galleries')
most_freq_words_dict.pop('Museums')
most_freq_words_dict2 = most_freq_words_dict.copy()

# create a wordcloud with the most frequent words
from wordcloud import WordCloud

wordcloud = WordCloud(width = 800, height = 800,
                background_color ='white',
                min_font_size = 10).generate_from_frequencies(most_freq_words_dict)

# plot the WordCloud image
plt.figure(figsize = (8, 8), facecolor = None)
plt.imshow(wordcloud)
plt.axis("off")
plt.tight_layout(pad = 0)

plt.show()

Hierarchical clustering using event description data¶

Next we explore exhibition descriptions using hierarchical clustering; a method used to group similar objects into clusters that follow a hierarchical structure. This can help conceptualise what a taxonomy of venue categories might look like. We use two methods to embed the data, term frequency-inverse document frequency (TF-IDF) and Google’s BERT.

The former is a statistical measure that evaluates how relevant a word is to a document in a collection of documents. The latter is a language model that can embed a sense of “context” to the textual data. This means that the model can quantitatively encode the meaning of a word based on the words that surround it.

Hierarchical clustering (tf-idf)¶

The first dendrogram shows the hierarchical clustering of the exhibition descriptions using TF-IDF. We cluster the data into seven groups. The x-axis provides a count (in brackets) of the exhibitition within each respective cluster. We also provide two sets of annotations. The annotations above the x-axis represent the most frequently occuring terms within the event description for each cluster. We limit this list to distinctive words. The annotations below the x-axis represent the most frequently occuring venues for each cluster.

import numpy as np
import seaborn as sns
import re
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import AgglomerativeClustering
from scipy.cluster.hierarchy import dendrogram, fcluster

# import nltk
# nltk.download('stopwords')
from nltk.corpus import stopwords

### collect all relevant textual data from the dataframe
# if description is not \n then append to slug2 in one line
dfs_rich['slug2'] = dfs_rich['title'].fillna('') + dfs_rich['description'].apply(lambda x: '' if x == '\n' else x)

### pre-process for NLP
# Load the documents and their corresponding categorical variables into a Pandas dataframe
df = pd.DataFrame({'text': dfs_rich['slug2'], 'category': dfs_rich['place_address']})

# summarise text for each unique place name
df['text'] = df.groupby('category')['text'].transform(lambda x: ' '.join(x))

#add new column with count for each category
df['cat_count'] = df.groupby('category')['category'].transform('count')
df.drop_duplicates(inplace=True)

# Clean the text
stop_words = set(stopwords.words('english'))

def clean_text(text):
    text = re.sub('[^a-zA-Z]', ' ', text)
    text = text.replace(',', '')
    text = text.lower().split()
    text = [word for word in text if word not in stop_words]
    text = ' '.join(text)
    return text

df = df[df['text'].notnull()]
df['clean_text'] = df['text'].apply(clean_text)

# Vectorize the text
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(df['clean_text'])


### Generate clusters

def get_most_common_word(df, more_words = [], no_of_words=25):
    # remove a list of all words that are not relevant
    # words_to_remove = [',','NSW','Sydney','Melbourne','Adelaide','South','SA','Brisbane','VIC',
    #                 '.', 'New', 'Australia', 'QLD', 'Vic', 'Wales', 'WA', 'Canberra', 'and',
    #                 'Perth', 'ACT', 'of', 'Qld', 'Victoria','Wollongong','TAS','Queensland','Newcastle',
    #                 'Street','Hobart','the','The','Launceston','Orange','UK','NT','London','USA',
    #                 'Paddington','Darwin','for','Western','Warrnambool','Ballarat','Northern','Territory',
    #                 'England','Watters','Macquarie','Artspace','St',"'s",'&','Potter','Kings','Ian','Cross',
    #                 '8','Llankelly','2011','Fremantle','Queen','Ivan','Dougherty','Tasmania','Central',
    #                 'Curtin','France','Tin','Sheds','York','Monash','Paris','Heide','Place','Vic.',]
    
    words_to_remove = []
    
    # add more words to the list of words to remove
    words_to_remove = words_to_remove + more_words

    all_words = []
    for i in df:
        for j in i.split(' '):
            all_words.append(j.replace(',',''))

    # find the most common words
    most_common_words = Counter(all_words).most_common(no_of_words)

    # remove the words from the list of most frequent words
    most_common_words = [word for word in most_common_words if word[0] not in words_to_remove]
    return most_common_words

def get_linkage_matrix(model, **kwargs):
    # Create linkage matrix and then plot the dendrogram

    # create the counts of samples under each node
    counts = np.zeros(model.children_.shape[0])
    n_samples = len(model.labels_)
    for i, merge in enumerate(model.children_):
        current_count = 0
        for child_idx in merge:
            if child_idx < n_samples:
                current_count += 1  # leaf node
            else:
                current_count += counts[child_idx - n_samples]
        counts[i] = current_count

    linkage_matrix = np.column_stack(
        [model.children_, model.distances_, counts]
    ).astype(float)

    return linkage_matrix

# # setting distance_threshold=0 ensures we compute the full tree.
# model_tfidf = AgglomerativeClustering(distance_threshold=0, n_clusters=None)
# model_tfidf = model_tfidf.fit(X.toarray())

# from textwrap import wrap

# cut = 7
# l_matrix = get_linkage_matrix(model_tfidf)
# df['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# # if a word is a duplicate then remove from all_words
# def find_duplicates(all_words, occurences=1):
#     duplicates = []
#     non_duplicates = []

#     for i in all_words:
#         if i in duplicates: continue
#         else:
#             if all_words.count(i) > occurences: duplicates.append(i)
#             else: non_duplicates.append(i)
#     return duplicates

# all_words = []

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.45), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.45), 
#                  xycoords='axes fraction', fontsize=9, color='red')
    
#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text']))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=2)
# all_words_to_remove.extend(['j','th','nd','exhibitionexhibited'])

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'],
#                                                                        more_words=all_words_to_remove)) if (idx < 5))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xycoords='axes fraction', fontsize=9)
    
#     annot2 = cluster_docs.sort_values('cat_count', ascending=False)['category'].values[0:3]
#     annot2 = '\n\n'.join(['\n'.join(wrap(line, 18)) for line in [i.split(',')[0] for i in annot2]])
#     # annot2 = '\n'.join(wrap(annot2, 18)) # breaks strings into new lines

#     plt.annotate(annot2, xy=(i/df['cluster'].nunique()-0.115, -0.24), 
#                 xytext=(i/df['cluster'].nunique()-0.115, -0.24), 
#                 xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - TF-IDF")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_tfidf.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_tfidf.png')

We provide the same visual however show the most frequent terms used within the place names as opposed to the event descriptions.

# cut = 7
# l_matrix = get_linkage_matrix(model_tfidf)
# df['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))

#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.45), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.45), 
#                  xycoords='axes fraction', fontsize=9, color='red')
    
#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['category']))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=3)

# all_words_to_remove.extend([',','NSW','Sydney','Melbourne','Adelaide','South','SA','Brisbane','VIC',
#                    '.', 'New', 'Australia', 'QLD', 'Vic', 'Wales', 'WA', 'Canberra', 'and',
#                    'Perth', 'ACT', 'of', 'Qld', 'Victoria','Wollongong','TAS','Queensland','Newcastle',
#                    'Street','Hobart','the','The','Launceston','Orange','UK','NT','London','USA',
#                    'Paddington','Darwin','for','Western','Warrnambool','Ballarat','Northern','Territory',
#                    'England','Watters','Macquarie','Artspace','St',"'s",'&','Potter','Kings','Ian','Cross',
#                    '8','Llankelly','2011','Fremantle','Queen','Ivan','Dougherty','Tasmania','Central',
#                    'Curtin','France','Tin','Sheds','York','Monash','Paris','Heide'])

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'],
#                                                                        more_words=all_words_to_remove)) if (idx < 5))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - TF-IDF")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_tfidf_placename.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_tfidf_placename.png')

Hierarchical clustering (BERT)¶

We repeat the methodology but now change the embedding method to BERT. We expect using BERT, a much more complex model, to better represent each venue. We find that the clusters are more widely populated, and output a different outcome.

# ### load bert model

# # !pip install transformers
# from transformers import BertTokenizer, BertModel
# from transformers import pipeline
# tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
# model = BertModel.from_pretrained("bert-base-uncased")

# ### encode text using bert (takes 2mins to run)
# import random

# # encode the text using bert
# def bert_encode(x):
#   encoded_input = tokenizer(x, return_tensors='pt')
#   output = model(**encoded_input)
#   return pd.DataFrame(output['pooler_output'].detach().numpy()).T

# # randomly sample 512 tokens from each row in df['clean_text']
# # some strings are smalle than 512
# df['clean_text_sampled'] = df['clean_text'].apply(lambda x: ' '.join(random.sample(x.split(' '), 340)) if len(x.split(' ')) >= 340 else x)
# X_bert = df['clean_text_sampled'].apply(lambda x: pd.Series(bert_encode([str(x)])[0]))

# # setting distance_threshold=0 ensures we compute the full tree.
# model_bert = AgglomerativeClustering(distance_threshold=0, n_clusters=None)
# model_bert = model_bert.fit(np.array(X_bert))

# cut = 7
# l_matrix = get_linkage_matrix(model_bert)
# df['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))

#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xycoords='axes fraction', fontsize=9, color='red')

#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text']))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=2)
# all_words_to_remove.extend(['j','th','nd','exhibitionexhibited','http','www','isbn'])

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'],
#                                                                        more_words=all_words_to_remove)) if (idx < 5))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.025), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.025), 
#                  xycoords='axes fraction', fontsize=9)
    
#     annot2 = cluster_docs.sort_values('cat_count', ascending=False)['category'].values[0:3]
#     annot2 = '\n\n'.join(['\n'.join(wrap(line, 18)) for line in [i.split(',')[0] for i in annot2]])
#     # annot2 = '\n'.join(wrap(annot2, 18)) # breaks strings into new lines

#     plt.annotate(annot2, xy=(i/df['cluster'].nunique()-0.115, -0.24), 
#                 xytext=(i/df['cluster'].nunique()-0.115, -0.24), 
#                 xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - BERT")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_bert.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert.png')

We provide the same visual however show the most frequent terms used within the place names as opposed to the event descriptions.

# cut = 7
# l_matrix = get_linkage_matrix(model_bert)
# df['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))

#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xycoords='axes fraction', fontsize=9, color='red')
    
#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['category'], no_of_words=25))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=4)
# all_words_to_remove.extend([',','NSW','Sydney','Melbourne','Adelaide','South','SA','Brisbane','VIC',
#                    '.', 'New', 'Australia', 'QLD', 'Vic','and','WA','Victoria','ACT','Qld',
#                    'of','Wollongong','TAS','Queensland','Newcastle',
#                    'Street','Hobart','the','The','Launceston','Orange','NT',
#                    'Paddington','Darwin','for','Western','Warrnambool','Ballarat','Northern','Territory',
#                    'England','Watters','Macquarie','Artspace','St',"'s",'&','Potter','Kings','Ian','Cross',
#                    '8','Llankelly','2011','Fremantle','Queen','Ivan','Dougherty','Tasmania','Central',
#                    'Curtin','France','Tin','Sheds','York','Monash','Heide',''])

# for i in df['cluster'].unique():
#     cluster_docs = df[df['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'],
#                                                                        more_words=all_words_to_remove)) if (idx < 5))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.05), 
#                  xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - BERT")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_bert_placename.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_placename.png')

Hierarchical clustering (BERT) after omitting erroneous geocodes that occur frequently¶

The next two dendograms are the same as above, however we omit the erroneous geocodes that occur frequently (the ones highlighted at the beginning of the notebook).

# # retrieve rows from most_freq_geocodes from a list of indices
# remove_these = most_freq_geocodes.head(20).iloc[most_freq_geocodes.head(20).index.isin([0,1,2,3,10,16])]

# # remove 5656 rows from df_daoo_relevant_events with same lat and long data from remove_these 
# dfs_rich['erroneous'] = np.where(dfs_rich['latitude'].isin(remove_these['latitude']) &\
#                                                 dfs_rich['longitude'].isin(remove_these['longitude']), True, False)

# ### pre-process for NLP
# # Load the documents and their corresponding categorical variables into a Pandas dataframe
# df2 = pd.DataFrame({'text': dfs_rich[dfs_rich['erroneous'] == 0]['slug2'], 
#                    'category': dfs_rich[dfs_rich['erroneous'] == 0]['place_address']})

# # summarise text for each unique place name
# df2['text'] = df2.groupby('category')['text'].transform(lambda x: ' '.join(x))

# #add new column with count for each category
# df2['cat_count'] = df2.groupby('category')['category'].transform('count')
# df2.drop_duplicates(inplace=True)

# # Clean the text
# stop_words = set(stopwords.words('english'))
# df2 = df2[df2['text'].notnull()]
# df2['clean_text'] = df2['text'].apply(clean_text)

# # randomly sample 512 tokens from each row in df['clean_text']
# # some strings are smalle than 512
# df2['clean_text_sampled'] = df2['clean_text'].apply(lambda x: ' '.join(random.sample(x.split(' '), 340)) if len(x.split(' ')) >= 340 else x)
# X_bert2 = df2['clean_text_sampled'].apply(lambda x: pd.Series(bert_encode([str(x)])[0]))

# # setting distance_threshold=0 ensures we compute the full tree.
# model_bert2 = AgglomerativeClustering(distance_threshold=0, n_clusters=None)
# model_bert2 = model_bert2.fit(np.array(X_bert2))

# cut = 7
# l_matrix = get_linkage_matrix(model_bert2)
# df2['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df2['cluster'].unique():
#     cluster_docs = df2[df2['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))

#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.15), 
#                  xycoords='axes fraction', fontsize=9, color='red')
    
#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text']))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=2)
# all_words_to_remove.extend(['j','p','r','h','c','k','th','nd','exhibitionexhibited','isbn','http','www'])

# for i in df2['cluster'].unique():
#     cluster_docs = df2[df2['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'],
#                                                                        more_words=all_words_to_remove)) if (idx < 5))
    
#     plt.annotate(annot, xy=(i/df2['cluster'].nunique()-0.1, 0.025), 
#                  xytext=(i/df2['cluster'].nunique()-0.1, 0.025), 
#                  xycoords='axes fraction', fontsize=9)
    
#     annot2 = cluster_docs.sort_values('cat_count', ascending=False)['category'].values[0:3]
#     annot2 = '\n\n'.join(['\n'.join(wrap(line, 18)) for line in [i.split(',')[0] for i in annot2]])
#     # annot2 = '\n'.join(wrap(annot2, 18)) # breaks strings into new lines

#     plt.annotate(annot2, xy=(i/df2['cluster'].nunique()-0.115, -0.24), 
#                 xytext=(i/df2['cluster'].nunique()-0.115, -0.24), 
#                 xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - BERT (after removing frequent errors)")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_bert2.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert2.png')

# cut = 7
# l_matrix = get_linkage_matrix(model_bert2)
# df2['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df2['cluster'].unique():
#     cluster_docs = df2[df2['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))

#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'])) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df['cluster'].nunique()-0.1, 0.175), 
#                  xytext=(i/df['cluster'].nunique()-0.1, 0.175), 
#                  xycoords='axes fraction', fontsize=9, color='red')

#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['category'], no_of_words=30))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=5)
# all_words_to_remove.extend([',','NSW','Sydney','Melbourne','Adelaide','South','SA','Brisbane','VIC',
#                    '.', 'New', 'QLD', 'Vic','and','WA','Victoria','ACT','Qld',
#                    'of','Wollongong','TAS','Queensland','Newcastle','Australia',
#                    'Street','Hobart','the','The','Launceston','Orange','NT',
#                    'Paddington','Darwin','for','Western','Warrnambool','Ballarat','Northern','Territory',
#                    'England','Watters','Macquarie','Artspace','St',"'s",'&','Potter','Kings','Ian','Cross',
#                    '8','Llankelly','2011','Fremantle','Queen','Ivan','Dougherty','Tasmania','Central',
#                    'Curtin','France','Tin','Sheds','York','Monash','Heide',''])

# for i in df2['cluster'].unique():
#     cluster_docs = df2[df2['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'],
#                                                                        more_words=all_words_to_remove)) if (idx < 20))
    
#     plt.annotate(annot, xy=(i/df2['cluster'].nunique()-0.1, 0.025), 
#                  xytext=(i/df2['cluster'].nunique()-0.1, 0.025), 
#                  xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - BERT (after removing frequent errors)")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

# # save the figure
# fig.savefig('images/daao_tlc/outputnew_bert2_placename.png', dpi=300, bbox_inches='tight')

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert2_placename.png')

Hierarchy Clustering - BERT by vicennium¶

Using the BERT model, we generate a series of dendrograms that show the hierarchical clustering of the exhibition descriptions by 20-year periods.

1900 - 1920¶

def temporal_dendos(startdt=1900, enddt=1920, cut_input=3):
    df_period = dfs_rich[(dfs_rich['decade_start'] >= startdt) & (dfs_rich['decade_start'] <= enddt)]
    ### pre-process for NLP
    # Load the documents and their corresponding categorical variables into a Pandas dataframe
    df_period = pd.DataFrame({'text': df_period['slug2'], 'category': df_period['place_address']})

    # summarise text for each unique place name
    df_period['text'] = df_period.groupby('category')['text'].transform(lambda x: ' '.join(x))

    #add new column with count for each category
    df_period['cat_count'] = df_period.groupby('category')['category'].transform('count')
    df_period.drop_duplicates(inplace=True)

    # Clean the text
    stop_words = set(stopwords.words('english'))
    df_period = df_period[df_period['text'].notnull()]
    df_period['clean_text'] = df_period['text'].apply(clean_text)

    # randomly sample 512 tokens from each row in df['clean_text']
    # some strings are smalle than 512
    df_period['clean_text_sampled'] = df_period['clean_text'].apply(lambda x: ' '.join(random.sample(x.split(' '), 300)) if len(x.split(' ')) >= 300 else x)
    X_bert_period = df_period['clean_text_sampled'].apply(lambda x: pd.Series(bert_encode([str(x)])[0]))

    # setting distance_threshold=0 ensures we compute the full tree.
    model_bert_period = AgglomerativeClustering(distance_threshold=0, n_clusters=None)
    model_bert_period = model_bert_period.fit(np.array(X_bert_period))

    ### generate dendrogram
    cut = cut_input
    l_matrix = get_linkage_matrix(model_bert_period)
    df_period['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
    dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

    all_words = []

    for i in df_period['cluster'].unique():
        cluster_docs = df_period[df_period['cluster'] == i]
        # print(i, get_most_common_word(cluster_docs['clean_text']))

        annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'])) if (idx < 3))
        
        plt.annotate(annot, xy=(i/df_period['cluster'].nunique()-0.1, 0.15), 
                    xytext=(i/df_period['cluster'].nunique()-0.1, 0.15), 
                    xycoords='axes fraction', fontsize=9, color='red')

        [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text']))]
        
    all_words_to_remove = find_duplicates(all_words, occurences=2)
    all_words_to_remove.extend(['j','th','nd','exhibitionexhibited','http','www','isbn'])

    for i in df_period['cluster'].unique():
        cluster_docs = df_period[df_period['cluster'] == i]
        # print(i, get_most_common_word(cluster_docs['clean_text']))
        annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['clean_text'],
                                                                        more_words=all_words_to_remove)) if (idx < 5))
        
        plt.annotate(annot, xy=(i/df_period['cluster'].nunique()-0.1, 0.025), 
                    xytext=(i/df_period['cluster'].nunique()-0.1, 0.025), 
                    xycoords='axes fraction', fontsize=9)
        
        annot2 = cluster_docs.sort_values('cat_count', ascending=False)['category'].values[0:3]
        annot2 = '\n\n'.join(['\n'.join(wrap(line, 18)) for line in [i.split(',')[0] for i in annot2]])
        # annot2 = '\n'.join(wrap(annot2, 18)) # breaks strings into new lines

        plt.annotate(annot2, xy=(i/df_period['cluster'].nunique()-0.115, -0.24), 
                    xytext=(i/df_period['cluster'].nunique()-0.115, -0.24), 
                    xycoords='axes fraction', fontsize=9)

    plt.title(f"Hierarchical Clustering Dendrogram - BERT - {startdt}-{enddt}")

    # make figure bigger
    fig = plt.gcf()
    fig.set_size_inches(14, 10)

    plt.show()

    # save the figure
    fig.savefig(f'images/daao_tlc/outputnew_bert_{startdt}_{enddt}.png', dpi=300, bbox_inches='tight')
# temporal_dendos(2000, 2030, cut_input=5)

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_1900_1920.png')

1920 - 1940¶

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_1920_1940.png')

1940 - 1960¶

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_1940_1960.png')

1960 - 1980¶

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_1960_1980.png')

1980 - 2000¶

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_1980_2000.png')

2000 - 2020¶

from IPython.display import Image
Image(filename='images/daao_tlc/outputnew_bert_2000_2030.png')

Organisations¶

Similar to venues, we first produce a word cloud for common terms used in organisation names. We find that the most frequent terms are “Society”, “Art”, “Gallery”, “Royal”, and “Arts”.

# bios = pd.read_csv('data/daao_biography_level_202202211719.csv')
# bios = bios[['_id','primary_name.family_name','primary_name.given_names']]
# bios[bios['_id'].str.contains('4e1e1424b846f2dae7001dc0')]

with open("data/daao_persongroup_level.json") as json_file: daoo_org_data=json.load(json_file)

# create a list of all the place names
all_place_names = pd.json_normalize(daoo_org_data)['primary_name'].astype(str).tolist()

# create a list of all the words in the place names
all_words = []
for place_name in all_place_names:
    all_words.extend(word_tokenize(place_name))

# find top 100 most frequent words
most_freq_words = Counter(all_words).most_common(40)

# remove a list of all words that are not relevant
words_to_remove = [',','.','and','of','the','The','for',"'s",'&','(',')','J.','Melbourne',
                   'UK','VIC','London',
                   'Jane','Lapham','*','Adelaide',"'",'Sydney','NSW','New','South','Wales']

# remove the words from the list of most frequent words
most_freq_words = [word for word in most_freq_words if word[0] not in words_to_remove]

most_freq_words_dict = dict(most_freq_words)

# # add value of two keys
# most_freq_words_dict['Gallery'] = most_freq_words_dict['Gallery'] + most_freq_words_dict['Galleries']
# most_freq_words_dict['Museum'] = most_freq_words_dict['Museum'] + most_freq_words_dict['Museums']

# # remove key 'Gallery'
# most_freq_words_dict.pop('Galleries')
# most_freq_words_dict.pop('Museums')

# create a wordcloud with the most frequent words
from wordcloud import WordCloud

wordcloud = WordCloud(width = 800, height = 800,
                background_color ='white',
                min_font_size = 10).generate_from_frequencies(most_freq_words_dict)

# plot the WordCloud image
plt.figure(figsize = (8, 8), facecolor = None)
plt.imshow(wordcloud)
plt.axis("off")
plt.tight_layout(pad = 0)

plt.show()

As shown above, the use of dendrograms can be an interesting exercise to explore the data. The initial idea was to repeat the same approach with organisations to assess any latent patterns through this taxonomic approrach. However, we find that the data is not as rich as the venue data and that there are many organisations with no biographies, no summaries, and no relations to people records, exhibition records, etc. Nevertheless we provide one dendrogram using organisation biographies.

Below is a list of the proportion of missing data for each field across DAAO organisations.

# find the proportion of null values in each column
print((pd.json_normalize(daoo_org_data).isnull().sum()/pd.json_normalize(daoo_org_data).shape[0])[0:20])

biography              0.869957
_cls                   0.000000
editing_complete       0.922318
is_deleted             0.000000
is_featured            0.000000
is_locked              0.000000
is_primary             0.000000
is_shadow              0.000000
primary_name           0.000858
periods_active         0.959227
record_status          0.000000
residences             0.959442
roles                  0.000000
sources                0.913519
slug                   0.000000
summary                0.917167
related_stub_people    0.998927
_id.$oid               0.000000
date_created.$date     0.024678
date_modified.$date    0.069528
dtype: float64

print((pd.json_normalize(daoo_org_data).isnull().sum()/pd.json_normalize(daoo_org_data).shape[0])[20:])

references                    0.896567
tags                          0.962017
type                          0.974464
types                         0.965665
related_events                0.972318
related_collections           0.985837
related_works                 0.922318
related_recognitions          0.996996
related_stub_person_groups    0.998927
copyright_agents              0.999571
urls                          0.896137
alternative_names             0.967811
locked_biographies            0.999356
other_occupations             0.996996
see_alsos                     0.982833
nla_id                        0.999356
source_database_ids           0.985837
dtype: float64

# from bs4 import BeautifulSoup

# orgs = pd.DataFrame(columns=['name','summary','bio'])

# # pd.json_normalize(daoo_org_data)['biography'].isnull().sum()/pd.json_normalize(daoo_org_data).shape[0] # 87% of rows have no biography

# for i,row in pd.json_normalize(daoo_org_data).iterrows():
#     try:
#         try: row['biography'][0]['text']
#         except: continue

#         # try: period_start = row['periods_active'][0]['start']['_date']
#         # except: period_start = None

#         # try: period_end = row['periods_active'][0]['end']['_date']
#         # except: period_end = None

#         # use pandas.concat to append new row to dataframe
#         orgs = pd.concat([orgs, pd.DataFrame({'name': [row['primary_name']],
#                                                 'summary': [row['summary']],
#                                                 'bio': [row['biography'][0]['text']],
#                                                 # 'period_start': [period_start],
#                                                 # 'period_end': [period_end]
#                                                 })], ignore_index=True)
        
#     except:
#         print(i)
#         break

# # remove empty summary
# orgs['summary'].fillna('', inplace=True)

# # remove rows with stub text
# orgs = orgs[~orgs['bio'].isin(orgs['bio'].value_counts().head(5).index.to_list())]

# # combine summary and bio
# orgs['bio'] = orgs['summary'] + orgs['bio'].apply(lambda x: BeautifulSoup(x, 'lxml').get_text())

# orgs = orgs[['name','bio']]

# ### encode text using bert (takes 4mins to run)

# ### pre-process for NLP
# # Load the documents and their corresponding categorical variables into a Pandas dataframe
# df_org = pd.DataFrame({'text': orgs['bio'], 'category': orgs['name']})

# #add new column with count for each category
# df_org['cat_count'] = df_org.groupby('category')['category'].transform('count')
# df_org.drop_duplicates(inplace=True)

# # Clean the text
# stop_words = set(stopwords.words('english'))

# def clean_text(text):
#     text = re.sub('[^a-zA-Z]', ' ', text)
#     text = text.replace(',', '')
#     text = text.lower().split()
#     text = [word for word in text if word not in stop_words]
#     text = ' '.join(text)
#     return text

# df_org['clean_text'] = df_org['text'].apply(clean_text)

# # randomly sample 512 tokens from each row in df['clean_text']
# # some strings are smalle than 512
# df_org['clean_text_sampled'] = df_org['clean_text'].apply(lambda x: ' '.join(random.sample(x.split(' '), 350)) if len(x.split(' ')) >= 350 else x)
# X_bert_org = df_org['clean_text_sampled'].apply(lambda x: pd.Series(bert_encode([str(x)])[0]))

# # setting distance_threshold=0 ensures we compute the full tree.
# model_bert_org = AgglomerativeClustering(distance_threshold=0, n_clusters=None)
# model_bert_org = model_bert_org.fit(np.array(X_bert_org))

Biographical data for each organisation (~130 organisations) is used to create the clusters. From here we show the most common terms used in the organisation name for each cluster.

# cut = 5
# l_matrix = get_linkage_matrix(model_bert_org)
# df_org['cluster'] = fcluster(l_matrix, cut, criterion='maxclust')
# dendrogram(l_matrix, orientation='top', truncate_mode="lastp", p=cut, show_leaf_counts=True)

# all_words = []

# for i in df_org['cluster'].unique():
#     cluster_docs = df_org[df_org['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     [all_words.append(i[0]) for idx,i in enumerate(get_most_common_word(cluster_docs['category'], no_of_words=5))]
    
# all_words_to_remove = find_duplicates(all_words, occurences=4)
# all_words_to_remove.extend(['h','co','e','test','Lapham','J.','Jane','','and','of','&','The','SymbioticA','Nevin','Smith'])

# for i in df_org['cluster'].unique():
#     cluster_docs = df_org[df_org['cluster'] == i]
#     # print(i, get_most_common_word(cluster_docs['clean_text']))
#     annot = "\n".join(i[0] for idx,i in enumerate(get_most_common_word(cluster_docs['category'], no_of_words=6,
#                                                                        more_words=all_words_to_remove)) if (idx < 3))
    
#     plt.annotate(annot, xy=(i/df_org['cluster'].nunique()-0.1, 0.05), 
#                  xytext=(i/df_org['cluster'].nunique()-0.1, 0.05), 
#                  xycoords='axes fraction', fontsize=12)
    
#     # annot2 = cluster_docs.sort_values('cat_count', ascending=False)['category'].values[0:3]
#     # annot2 = '\n\n'.join(['\n'.join(wrap(line, 18)) for line in [i.split(',')[0] for i in annot2]])
#     # # annot2 = '\n'.join(wrap(annot2, 18)) # breaks strings into new lines

#     # plt.annotate(annot2, xy=(i/df_org['cluster'].nunique()-0.115, -0.24), 
#     #             xytext=(i/df_org['cluster'].nunique()-0.115, -0.24), 
#     #             xycoords='axes fraction', fontsize=9)

# plt.title("Hierarchical Clustering Dendrogram - BERT, Organisations")

# # make figure bigger
# fig = plt.gcf()
# fig.set_size_inches(14, 10)

# plt.show()

from IPython.display import Image
Image(filename='images/daao_tlc/output5.png')

Temporal analysis of venues¶

To further explore frequently used terms in place names, we produce an interactive time series of the number of events per decade by associated terms.

Before the visuals, we provide an ordered list of the decades with the most event activity, and also a list of the top 20 most frequent terms used in place names.

events_df = dfs_rich.copy()

# create a new column with the decade of start year
events_df['decade'] = events_df['start_year'].apply(lambda x: str(x)[:3] + '0')
events_df['decade'] = events_df['decade'].apply(lambda x: '2000' if x == '2020' else x)
events_df['decade'] = events_df['decade'].astype(int)
events_df['decade'].value_counts()

  2286
  2272
  1413
   651
   526
   159
    90
    41
    40
    39
    33
    31
Name: decade, dtype: int64

most_freq_words_dict2

{'Gallery': 3929,
 'Art': 3404,
 'Centre': 665,
 'Museum': 775,
 'University': 627,
 'Contemporary': 594,
 'Foundation': 431,
 'Experimental': 429,
 'City': 367,
 'National': 328,
 'Modern': 318,
 'Regional': 283,
 'Institute': 254,
 'Arts': 229,
 'Australian': 197,
 'Fine': 184,
 'Place': 121,
 'Region': 108,
 'School': 75,
 'Hall': 68,
 'Space': 63,
 'College': 58,
 'House': 57}

words_by_decade = pd.DataFrame()

for key in most_freq_words_dict2:
    words_this_decade = events_df[events_df['place_address'].str.contains(key, na=False)]\
                        .groupby('decade')['place_address']\
                        .count()\
                        .reset_index(name='count')\
                        .sort_values(['decade'], ascending=True)
    words_this_decade['term'] = key
    words_by_decade = pd.concat([words_by_decade, words_this_decade], ignore_index=True)

import plotly.express as px
fig = px.line(words_by_decade, x="decade", y="count", color='term',
                title='Count of each term by decade')

# # make figure size bigger
# # fig.update_layout(
# #     autosize=False,
# #     width=700,
# #     height=500,
# # )

fig.show()

# events_df['diff'] = events_df['end_year'] - events_df['start_year']
# # events_df[(events_df['place_name'].str.contains('Contemporary')) & (events_df['diff'] > 1)]

# using start_date, end_date, place_name, calculate the difference between start_date of the first event in an venue and end_date of the last event in that venue
places_with_start_end = []
[places_with_start_end.extend([p, events_df.loc[events_df['place_address'] == p, 'start_year'].min(), 
                              events_df.loc[events_df['place_address'] == p, 'end_year'].max()] for p in events_df['place_address'].unique())]

places_with_start_end = pd.DataFrame(places_with_start_end)
places_with_start_end.columns = ['place_address', 'start_year', 'end_year']
places_with_start_end['diff'] = places_with_start_end['end_year'] - places_with_start_end['start_year']
places_with_start_end = places_with_start_end[places_with_start_end['diff'] >= 0]
# places_with_start_end.sort_values('diff', ascending=False)

Beyond the above time series, we also provide a series of bar charts, however instead of illustrating event activity within a certain decade, we plot the number of venues active across time. We calculate the start and end of a venue’s “lifespan” by using the earliest and latest event date associated with a venue. We do this by filtering on venues with associated terms in the place name.

Most bar charts follow the same upwards trends shown in the above time series, however some terms that highlight some different trajectories consist of “contemporary”, “society”, “experimental” and “space”.

decade_placeholder = pd.DataFrame([0] * 13, index=range(1900, 2020+1, 10)).reset_index().rename(columns={'index':'decade', 0:'count'})
decade_placeholder['decade'] = decade_placeholder['decade'].astype(str)

for term in most_freq_words_dict2:
    # create a wide form of contemporary with each decade as a column and a binary value for whether the venue existed in that decade
    # use start_year as the start and end_year as the end of the venue
    contemporary = places_with_start_end[(places_with_start_end['place_address'].str.contains(term))].sort_values('start_year', ascending=True)
    contemporary_wide = pd.DataFrame()

    for i,row in contemporary.iterrows():
        for year in range(int(row['start_year']), int(row['end_year'])+1): 
            contemporary_wide.loc[row['place_address'], str(year)[:3] + '0'] = 1

    contemporary_wide = contemporary_wide.\
        fillna(0).\
        reset_index().\
        rename(columns={'index':'place_address'}).\
        sum().tail(-1).\
        reset_index().\
        rename(columns={'index':'decade', 0:'count'})
    
    contemporary_wide = pd.merge(contemporary_wide, decade_placeholder, on='decade', how='outer')
    contemporary_wide['count'] = contemporary_wide['count_x'] + contemporary_wide['count_y']
    contemporary_wide = contemporary_wide[['decade', 'count']].sort_values('decade', ascending=True)

    fig = px.bar(contemporary_wide, x="decade", y='count',
                    title=f'Number of venues that existed in each decade, {term}')

    # remove y-axis label
    fig.update_yaxes(title_text='')
    fig.update_xaxes(title_text='')

    # # make figure size bigger
    # fig.update_layout(
    #     autosize=False,
    #     width=700,
    #     height=400,
    # )

    fig.show()

Proportions¶

We repeat the same temporal visuals as above, however this time we plot the proportions to effectively normalise the data with respect to the number of events across decades, and the number of venues that existed across time, repsectively.

Note that the y-axis limits in the bar charts changes for each term.

words_by_decade_prop = []

# find the proportions of each word in each decade
for i in most_freq_words_dict2.keys():
    for j in events_df['decade'].unique():
        prop = events_df[(events_df['place_address'].str.contains(i)) & (events_df['decade'] == j)].shape[0] /events_df[events_df['decade'] == j].shape[0]
        words_by_decade_prop.append([i,j,prop])
    
words_by_decade_prop = pd.DataFrame(words_by_decade_prop)
words_by_decade_prop.columns = ['word','decade','proportion']

# plot proportion of each term by decade
# make it interactive

import plotly.express as px

fig = px.line(words_by_decade_prop.sort_values('decade'), x="decade", y="proportion", color='word',
                title='Proportion of each term by decade')

# make figure size bigger
# fig.update_layout(
#     autosize=False,
#     width=700,
#     height=500,
# )

fig.show()

allplaces_wide = pd.DataFrame()

for i,row in places_with_start_end.iterrows():
    for year in range(int(row['start_year']), int(row['end_year'])+1): 
        allplaces_wide.loc[row['place_address'], str(year)[:3] + '0'] = 1

decade_placeholder_forprop = allplaces_wide.\
                            fillna(0).\
                            reset_index().\
                            rename(columns={'index':'place_address'}).\
                            sum().tail(-1).\
                            reset_index().\
                            rename(columns={'index':'decade', 0:'count'}).\
                            sort_values('decade', ascending=True)

for term in most_freq_words_dict2:
    # create a wide form of contemporary with each decade as a column and a binary value for whether the venue existed in that decade
    # use start_year as the start and end_year as the end of the venue
    contemporary = places_with_start_end[(places_with_start_end['place_address'].str.contains(term))].sort_values('start_year', ascending=True)
    contemporary_wide = pd.DataFrame()

    for i,row in contemporary.iterrows():
        for year in range(int(row['start_year']), int(row['end_year'])+1): 
            contemporary_wide.loc[row['place_address'], str(year)[:3] + '0'] = 1

    contemporary_wide = contemporary_wide.\
        fillna(0).\
        reset_index().\
        rename(columns={'index':'place_address'}).\
        sum().tail(-1).\
        reset_index().\
        rename(columns={'index':'decade', 0:'count'})
    
    contemporary_wide = pd.merge(contemporary_wide, decade_placeholder_forprop, on='decade', how='outer')
    contemporary_wide['prop'] = np.where(contemporary_wide['count_x'] > 0, 
                                     contemporary_wide['count_x']/contemporary_wide['count_y'], 0)
    contemporary_wide = contemporary_wide[['decade', 'prop']].sort_values('decade', ascending=True)

    fig = px.bar(contemporary_wide, x="decade", y='prop',
                    title=f'Number of venues that existed in each decade, {term}')

    # remove y-axis label
    fig.update_yaxes(title_text='')
    fig.update_xaxes(title_text='')

    # make figure size bigger
    # fig.update_layout(
    #     autosize=False,
    #     width=700,
    #     height=400,
    # )

    fig.show()

DAAO venues and organisations

Contents